Predictions and error of a line

Before training anything, use a given line: predict each example and measure how wrong the line is.

What to build

A program that applies a line to some examples and prints its mean squared error.

Requirements

  • Print the predictions with 2 decimal places, separated by spaces.
  • Then print MSE M with 3 decimal places.

Input

Line 1: w and b. Line 2: the x values. Line 3: the true y values.

Output

Two lines.

Constraints

1 to 1000 examples.

Example

Input:

2 1
1 2 3
3 5 8

Output:

3.00 5.00 7.00
MSE 0.333

Why it works

The mean squared error turns all the residuals into one number, so lines can be compared: training looks for the line that makes it smallest.

Examples

Example 1 · Input
2 1
1 2 3
3 5 8
Expected output
3.00 5.00 7.00
MSE 0.333
Example 2 · Input
0.5 0
2 4
1 1
Expected output
1.00 2.00
MSE 0.500

Prompts inside input() are ignored when checking, so input("n: ") is fine.

Hints

Hint 1

Each prediction is w × x + b.

Hint 2

The MSE is the mean of (prediction − y)².

Solution

Show one correct solution (try it yourself first)
w, b = map(float, input().split())
xs = list(map(float, input().split()))
ys = list(map(float, input().split()))
predictions = [w * x + b for x in xs]
mse = sum((p - y) ** 2 for p, y in zip(predictions, ys)) / len(xs)
print(*[f"{p:.2f}" for p in predictions])
print(f"MSE {mse:.3f}")

Lesson: Linear regression →